Hi Folks,
Cuis 3.2 is available at http://www.jvuletich.org/Cuis/Index.html . From the release notes: New in Cuis 3.2 ------------------- * Enhanced look for menus and several other widgets * TIFFReader (100% Smalltalk code) * Many fixes and code enhancements and cleanup, especially in Morphic (World, Canvas, DamageRecorder, etc) * Text / TextAttribute cleanup Many of the changes will be of interest to those working on updating integrating SimpleMorphic into Squeak and Pharo. As usual, comments, discussion, bug reports, kudos and donations are all welcome! Cheers, Juan Vuletich |
On 4/12/11 5:15 PM, "Juan Vuletich" <[hidden email]> wrote: > Hi Folks, > > Cuis 3.2 is available at http://www.jvuletich.org/Cuis/Index.html . From > the release notes: > > New in Cuis 3.2 > ------------------- > * Enhanced look for menus and several other widgets > * TIFFReader (100% Smalltalk code) > * Many fixes and code enhancements and cleanup, especially in > Morphic (World, Canvas, DamageRecorder, etc) > * Text / TextAttribute cleanup > > Many of the changes will be of interest to those working on updating > integrating SimpleMorphic into Squeak and Pharo. > > As usual, comments, discussion, bug reports, kudos and donations are all > welcome! > > Cheers, > Juan Vuletich Just downloading it.... Edgar |
In reply to this post by Juan Vuletich-4
On 4/12/11 5:15 PM, "Juan Vuletich" <[hidden email]> wrote: > Hi Folks, > > Cuis 3.2 is available at http://www.jvuletich.org/Cuis/Index.html . From > the release notes: > > New in Cuis 3.2 > ------------------- > * Enhanced look for menus and several other widgets > * TIFFReader (100% Smalltalk code) > * Many fixes and code enhancements and cleanup, especially in > Morphic (World, Canvas, DamageRecorder, etc) > * Text / TextAttribute cleanup > > Many of the changes will be of interest to those working on updating > integrating SimpleMorphic into Squeak and Pharo. > > As usual, comments, discussion, bug reports, kudos and donations are all > welcome! > > Cheers, > Juan Vuletich Like the LayoutMorphTest testLayout1 etc, this drive me to 'Morphic-Layouts' classes and wonder why this is not on SimpleMorphic ..... And a question. Why not a ToolBuilder in Cuis, you think is unneeded ? I ask because in the future I would like to port some Trunk tools to Cuis. Edgar |
Edgar J. De Cleene wrote:
> > On 4/12/11 5:15 PM, "Juan Vuletich" <[hidden email]> wrote: > > >> Hi Folks, >> >> Cuis 3.2 is available at http://www.jvuletich.org/Cuis/Index.html . From >> the release notes: >> >> New in Cuis 3.2 >> ------------------- >> * Enhanced look for menus and several other widgets >> * TIFFReader (100% Smalltalk code) >> * Many fixes and code enhancements and cleanup, especially in >> Morphic (World, Canvas, DamageRecorder, etc) >> * Text / TextAttribute cleanup >> >> Many of the changes will be of interest to those working on updating >> integrating SimpleMorphic into Squeak and Pharo. >> >> As usual, comments, discussion, bug reports, kudos and donations are all >> welcome! >> >> Cheers, >> Juan Vuletich >> > > > Like the LayoutMorphTest testLayout1 etc, this drive me to 'Morphic-Layouts' > classes and wonder why this is not on SimpleMorphic ..... > Well, I did all this after SimpleMorphic. It was one of the big news of Cuis 3.1. I suggest updating the code in SimpleMorphic to be on par with the latest on Cuis. Besides the much nicer look, you also get better performance and simpler code in _many_ places. > And a question. > > Why not a ToolBuilder in Cuis, you think is unneeded ? I ask because in the > future I would like to port some Trunk tools to Cuis. > Just because I didn't need it yet. Please go ahead and build it. It will be a great way to ease porting stuff from Squeak. Cheers, Juan Vuletich > Edgar > |
In reply to this post by Juan Vuletich-4
Juan the new menus look great!
On Apr 12, 2011, at 1:15 PM, Juan Vuletich <[hidden email]> wrote: > Hi Folks, > > Cuis 3.2 is available at http://www.jvuletich.org/Cuis/Index.html . From the release notes: > > New in Cuis 3.2 > ------------------- > * Enhanced look for menus and several other widgets > * TIFFReader (100% Smalltalk code) > * Many fixes and code enhancements and cleanup, especially in Morphic (World, Canvas, DamageRecorder, etc) > * Text / TextAttribute cleanup > > Many of the changes will be of interest to those working on updating integrating SimpleMorphic into Squeak and Pharo. > > As usual, comments, discussion, bug reports, kudos and donations are all welcome! > > Cheers, > Juan Vuletich > |
In reply to this post by Juan Vuletich-4
Hi,
Today I restarted my efforts of basing SimpleMorphic on Cuis' morphic. My previous try stalled, because it proved to be hard to gap the divergence of Squeak and Cuis outside the code I brought from there. (IIRC I was fiddling with CharacterScanner incompatibilities lately) So my new approach is to sever a bigger chunk from Cuis .  - I grabbed not only Morphic and the Theme categories but also all the code from System-Text, Tools, Shout and Graphics-Text and three more classes (Preferences Preference RunArray). [1]  - Then loaded it with the little utility I attached [2]. This converts category and class names and class references.  - Then I've loaded the SMx package, and started the long battle with the errors popping up, eventually slipping into changing the extracted SimpleMorphic package. I've uploaded the results to the SimpleMorphicSqueak repo [3]. Hopefully it can be loaded into a clean trunk image. It's possible to create a new project, open the world menu and return to the previous project. The next steps will be to make the tools (from Cuis) work, by improving the SMx package. I hope this will solder the code into the Trunk. Then make the ToolBuilder work, and then we hopefully can pick up the thread where we left and merge with Edgar. What do you think? Balázs p.s: We have 3 overrides in Trunk code. Two in Rectangle (Rectangle >> quickMerge:, Rectangle class >> merging: ). Both to accept nil instead of Rectangles. The third is DisplayScreen >> defaultCanvasClass. It dispatches on Project current to get the canvas class, so SimpleMorphic can use its own canvas. Maybe they can integrated into the Trunk. [1] This is the code I use to get the code out from Cuis: fs := FileStream newFileNamed: 'SimpleMorphic-3.2.st'. #('Morphic' 'Theme' 'System-Text' 'Tools' 'Shout' 'Graphics-Text') do: [ :category |     (SystemOrganization categoriesMatching: category, '*') do: [ :each |         SystemOrganization fileOutCategory: each on: fs ] ]. #(Preferences Preference RunArray) do: [ :className | (Smalltalk classNamed: className) fileOutOn: fs ]. fs close. [2] This is the code to load the filed out code: transformer := SMxFilePackageTransformer fromFileNamed: 'SimpleMorphic-3.2.st'. transformer filePackage fixClassOrder. transformer filePackage fileIn [3] Load script: Installer squeaksource     project: 'SimpleMorphicSqueak';     install: 'SimpleMorphic-kb.8';     install: 'SMx-kb.32' SMx-Code Loader.st (6K) Download Attachment |
On 4/14/11 2:38 AM, "Balázs Kósi" <[hidden email]> wrote: > Hi, > > Today I restarted my efforts of basing SimpleMorphic on Cuis' morphic. > My previous try stalled, because it proved to be hard to gap the > divergence of Squeak and Cuis outside the code I brought from there. > (IIRC I was fiddling with CharacterScanner incompatibilities lately) > > So my new approach is to sever a bigger chunk from Cuis . >  - I grabbed not only Morphic and the Theme categories but also all > the code from System-Text, Tools, Shout and Graphics-Text and three > more classes (Preferences Preference RunArray). [1] >  - Then loaded it with the little utility I attached [2]. This > converts category and class names and class references. >  - Then I've loaded the SMx package, and started the long battle with > the errors popping up, eventually slipping into changing the extracted > SimpleMorphic package. > > I've uploaded the results to the SimpleMorphicSqueak repo [3]. > Hopefully it can be loaded into a clean trunk image. It's possible to > create a new project, open the world menu and return to the previous > project. > > The next steps will be to make the tools (from Cuis) work, by > improving the SMx package. I hope this will solder the code into the > Trunk. Then make the ToolBuilder work, and then we hopefully can pick > up the thread where we left and merge with Edgar. What do you think? > > Balázs > > p.s: We have 3 overrides in Trunk code. Two in Rectangle (Rectangle >> > quickMerge:, Rectangle class >> merging: ). Both to accept nil instead > of Rectangles. The third is DisplayScreen >> defaultCanvasClass. It > dispatches on > Project current to get the canvas class, so SimpleMorphic can use its > own canvas. Maybe they can integrated into the Trunk. > > [1] This is the code I use to get the code out from Cuis: > > fs := FileStream newFileNamed: 'SimpleMorphic-3.2.st'. > #('Morphic' 'Theme' 'System-Text' 'Tools' 'Shout' 'Graphics-Text') do: > [ :category | >     (SystemOrganization categoriesMatching: category, '*') do: [ :each | >         SystemOrganization fileOutCategory: each on: fs ] ]. > #(Preferences Preference RunArray) do: [ :className | (Smalltalk > classNamed: className) fileOutOn: fs ]. > fs close. > > [2] This is the code to load the filed out code: > > transformer := SMxFilePackageTransformer fromFileNamed: > 'SimpleMorphic-3.2.st'. > transformer filePackage fixClassOrder. > transformer filePackage fileIn > > [3] Load script: > > Installer squeaksource >     project: 'SimpleMorphicSqueak'; >     install: 'SimpleMorphic-kb.8'; >     install: 'SMx-kb.32' Excellent! Just trying with fresh Squeak4.3-11157 and update to 11241. Gives some dnu, but nothing that can not be fixed Today can't do much work , but try to focus in get rid of MCTool dependencies for doing a merge between your code and mine. For this I should move ToolBuilder-Kernel-edc.48, ToolBuilder-Morphic-edc.77 and http://source.squeak.org/inbox/Monticello-edc.457.mcz to Trunk first . Agree ? Edgar |
On 14.04.2011, at 12:32, Edgar J. De Cleene wrote: > > > > On 4/14/11 2:38 AM, "Balázs Kósi" <[hidden email]> wrote: > >> Hi, >> >> Today I restarted my efforts of basing SimpleMorphic on Cuis' morphic. >> My previous try stalled, because it proved to be hard to gap the >> divergence of Squeak and Cuis outside the code I brought from there. >> (IIRC I was fiddling with CharacterScanner incompatibilities lately) >> >> So my new approach is to sever a bigger chunk from Cuis . >> - I grabbed not only Morphic and the Theme categories but also all >> the code from System-Text, Tools, Shout and Graphics-Text and three >> more classes (Preferences Preference RunArray). [1] >> - Then loaded it with the little utility I attached [2]. This >> converts category and class names and class references. >> - Then I've loaded the SMx package, and started the long battle with >> the errors popping up, eventually slipping into changing the extracted >> SimpleMorphic package. >> >> I've uploaded the results to the SimpleMorphicSqueak repo [3]. >> Hopefully it can be loaded into a clean trunk image. It's possible to >> create a new project, open the world menu and return to the previous >> project. >> >> The next steps will be to make the tools (from Cuis) work, by >> improving the SMx package. I hope this will solder the code into the >> Trunk. Then make the ToolBuilder work, and then we hopefully can pick >> up the thread where we left and merge with Edgar. What do you think? >> >> Balázs >> >> p.s: We have 3 overrides in Trunk code. Two in Rectangle (Rectangle >> >> quickMerge:, Rectangle class >> merging: ). Both to accept nil instead >> of Rectangles. The third is DisplayScreen >> defaultCanvasClass. It >> dispatches on >> Project current to get the canvas class, so SimpleMorphic can use its >> own canvas. Maybe they can integrated into the Trunk. >> >> [1] This is the code I use to get the code out from Cuis: >> >> fs := FileStream newFileNamed: 'SimpleMorphic-3.2.st'. >> #('Morphic' 'Theme' 'System-Text' 'Tools' 'Shout' 'Graphics-Text') do: >> [ :category | >> (SystemOrganization categoriesMatching: category, '*') do: [ :each | >> SystemOrganization fileOutCategory: each on: fs ] ]. >> #(Preferences Preference RunArray) do: [ :className | (Smalltalk >> classNamed: className) fileOutOn: fs ]. >> fs close. >> >> [2] This is the code to load the filed out code: >> >> transformer := SMxFilePackageTransformer fromFileNamed: >> 'SimpleMorphic-3.2.st'. >> transformer filePackage fixClassOrder. >> transformer filePackage fileIn >> >> [3] Load script: >> >> Installer squeaksource >> project: 'SimpleMorphicSqueak'; >> install: 'SimpleMorphic-kb.8'; >> install: 'SMx-kb.32' > > > Excellent! > Just trying with fresh Squeak4.3-11157 and update to 11241. > > > Gives some dnu, but nothing that can not be fixed > Today can't do much work , but try to focus in get rid of MCTool > dependencies for doing a merge between your code and mine. Sounds good. > For this I should move ToolBuilder-Kernel-edc.48, ToolBuilder-Morphic-edc.77 > and http://source.squeak.org/inbox/Monticello-edc.457.mcz to Trunk first . > > Agree ? No, there are problems with your code. See my other message in the "Monticello-edc.456" thread. - Bert - |
In reply to this post by Balázs Kósi
On 4/14/11, Balázs Kósi <[hidden email]> wrote:
> Hi, > > Today I restarted my efforts of basing SimpleMorphic on Cuis' morphic. > My previous try stalled, because it proved to be hard to gap the > divergence of Squeak and Cuis outside the code I brought from there. > (IIRC I was fiddling with CharacterScanner incompatibilities lately) > > So my new approach is to sever a bigger chunk from Cuis . > - I grabbed not only Morphic and the Theme categories but also all > the code from System-Text, Tools, Shout and Graphics-Text and three > more classes (Preferences Preference RunArray). [1] > - Then loaded it with the little utility I attached [2]. This > converts category and class names and class references. > - Then I've loaded the SMx package, and started the long battle with > the errors popping up, eventually slipping into changing the extracted > SimpleMorphic package. > > I've uploaded the results to the SimpleMorphicSqueak repo [3]. > Hopefully it can be loaded into a clean trunk image. It's possible to > create a new project, open the world menu and return to the previous > project. > > The next steps will be to make the tools (from Cuis) work, by > improving the SMx package. I hope this will solder the code into the > Trunk. Then make the ToolBuilder work, and then we hopefully can pick > up the thread where we left and merge with Edgar. What do you think? > > Balázs > > p.s: We have 3 overrides in Trunk code. Two in Rectangle (Rectangle >> > quickMerge:, Rectangle class >> merging: ). Both to accept nil instead > of Rectangles. The third is DisplayScreen >> defaultCanvasClass. It > dispatches on > Project current to get the canvas class, so SimpleMorphic can use its > own canvas. Maybe they can integrated into the Trunk. > > [1] This is the code I use to get the code out from Cuis: > > fs := FileStream newFileNamed: 'SimpleMorphic-3.2.st'. > #('Morphic' 'Theme' 'System-Text' 'Tools' 'Shout' 'Graphics-Text') do: > [ :category | > (SystemOrganization categoriesMatching: category, '*') do: [ :each | > SystemOrganization fileOutCategory: each on: fs ] ]. > #(Preferences Preference RunArray) do: [ :className | (Smalltalk > classNamed: className) fileOutOn: fs ]. > fs close. > > [2] This is the code to load the filed out code: > > transformer := SMxFilePackageTransformer fromFileNamed: > 'SimpleMorphic-3.2.st'. > transformer filePackage fixClassOrder. > transformer filePackage fileIn > [3] Load script: > > Installer squeaksource > project: 'SimpleMorphicSqueak'; > install: 'SimpleMorphic-kb.8'; > install: 'SMx-kb.32' > Thank you Balázs for this description which shows with enough details what you did. I used your load script [3] It loaded fine into Balázs Squeak 4.3 alphta -11241 (trunk) as of today. With menu 'Project / New Project' --> 'SMx Morphic Project' I can create a SimpleMorphic project enter it and leave it. In addition I can open a SystemBrowser. This is a good milestone. The GUI looks nice as the CUIS 3.2 I checked out yesterday. Cuis 3.2 has Morph allSubclasses size 52 --------------------------------------------------------------------- Now for the things which do not work yet If I 'right-click' on the category pane to bring up the find class command results in a debugger window. Opening a workspace brings the same result. --------------------------------------------------------------------- UIManager --------------------------------------------------------------------- I see that you have UIManager subclass: #SimpleMorphicUIManager --------------------------------------------------------------------- Questions --------------------------------------------------------------------- 1. Looking at the class category I see the prefix SimpleMorphic and SMx How do they differ? 2. How far have you got into looking at what Edgar did? He writes that he can bring up the Monticello browser though with some hacks (see other thread; Bert and Andreas dont' want an abstract method signature on UIManager which includes aWorld as a parameter. Regards --Hannes |
In reply to this post by Bert Freudenberg
On 4/14/11 8:09 AM, "Bert Freudenberg" <[hidden email]> wrote: > No, there are problems with your code. See my other message in the > "Monticello-edc.456" thread. > > - Bert - Well, now i know this is wrong. Like know what is right. Usinhg DependencyBrowser in fresh image MCTool>>#buttonRow: We have AlignmentMorph and PluggableButtonMorph MCWorkingCopyBrowser>>#editLoadScripts We have MenuMorph MCTool>>#multiListMorph:selection:listSelection:menu: We have PluggableListMorphOfMany And we could follow, but the critics was for MCRepository class>>#fillInTheBlankConfigure: I copy the code fillInTheBlankConfigure: aTemplateString | chunk repo | aTemplateString ifNil: [ ^ false ]. chunk := FillInTheBlankMorph request: self fillInTheBlankRequest initialAnswer: aTemplateString centerAt: Sensor cursorPoint inWorld: World onCancelReturn: nil acceptOnCR: false answerExtent: 400@120. chunk ifNotNil: [ repo := self readFrom: chunk readStream. repo creationTemplate: chunk. ]. ^ repo Is this best as my solution? Was in place for four years .... stephaneducasse 2/4/2006 20:47 Edgar |
Edgar
Do you mean what is shown in the screen shot? --Hannes On 4/14/11, Edgar J. De Cleene <[hidden email]> wrote: > > > > On 4/14/11 8:09 AM, "Bert Freudenberg" <[hidden email]> wrote: > >> No, there are problems with your code. See my other message in the >> "Monticello-edc.456" thread. >> >> - Bert - > > Well, now i know this is wrong. > Like know what is right. > > Usinhg DependencyBrowser in fresh image > > MCTool>>#buttonRow: We have AlignmentMorph and PluggableButtonMorph > MCWorkingCopyBrowser>>#editLoadScripts We have MenuMorph > MCTool>>#multiListMorph:selection:listSelection:menu: We have > PluggableListMorphOfMany > > And we could follow, but the critics was for > MCRepository class>>#fillInTheBlankConfigure: > > I copy the code > > fillInTheBlankConfigure: aTemplateString > | chunk repo | > > aTemplateString ifNil: [ ^ false ]. > chunk := FillInTheBlankMorph > request: self fillInTheBlankRequest > initialAnswer: aTemplateString > centerAt: Sensor cursorPoint > inWorld: World > onCancelReturn: nil > acceptOnCR: false > answerExtent: 400@120. > > chunk > ifNotNil: [ > repo := self readFrom: chunk readStream. > repo creationTemplate: chunk. > ]. > > ^ repo > > > Is this best as my solution? > Was in place for four years .... stephaneducasse 2/4/2006 20:47 > > Edgar > > > > DependencyBrowserPackagesDependantOnMonticello.PNG (43K) Download Attachment |
In reply to this post by Hannes Hirzel
>
> 1. Looking at the class category I see the prefix > > SimpleMorphic > and > SMx > SimpleMorphic is what comes from Cuis and SMx contains the classes which have to be in Squeak to accomodate SimpleMorphic. Name: SMx-dtl.6 Author: dtl Time: 24 February 2011, 1:18:35.649 pm UUID: da150bdb-5ebf-405e-a303-bb8e2093442e Ancestors: SMx-dtl.5 Move all Squeak extensions from package SimpleMorphic to package SMx in order to keep package SimpleMorphic as close as possible to the base distribution. |
In reply to this post by Edgar De Cleene
On 14.04.2011, at 17:15, Edgar J. De Cleene wrote: > On 4/14/11 8:09 AM, "Bert Freudenberg" <[hidden email]> wrote: > >> No, there are problems with your code. See my other message in the >> "Monticello-edc.456" thread. >> >> - Bert - > > Well, now i know this is wrong. > Like know what is right. > > Usinhg DependencyBrowser in fresh image > > MCTool>>#buttonRow: We have AlignmentMorph and PluggableButtonMorph > MCWorkingCopyBrowser>>#editLoadScripts We have MenuMorph > MCTool>>#multiListMorph:selection:listSelection:menu: We have > PluggableListMorphOfMany > > And we could follow, but the critics was for > MCRepository class>>#fillInTheBlankConfigure: > > I copy the code > > fillInTheBlankConfigure: aTemplateString > | chunk repo | > > aTemplateString ifNil: [ ^ false ]. > chunk := FillInTheBlankMorph > request: self fillInTheBlankRequest > initialAnswer: aTemplateString > centerAt: Sensor cursorPoint > inWorld: World > onCancelReturn: nil > acceptOnCR: false > answerExtent: 400@120. > > chunk > ifNotNil: [ > repo := self readFrom: chunk readStream. > repo creationTemplate: chunk. > ]. > > ^ repo IMHO the right way would be to simply use the existing multiLineRequest:centerAt:initialAnswer:answerHeight: method. Only if it turns out there is a need to specify the actual extent, and not just the height, a multiLineRequest:centerAt:initialAnswer:answerExtent: method could be added. The more methods we add to ToolBuilder, the harder it is to implement a new graphics framework. So we should take some care with this. - Bert - |
In reply to this post by Hannes Hirzel
El 14/04/11 12:28, "Hannes Hirzel" <[hidden email]> escribió: > Edgar > > > Do you mean what is shown in the screen shot? > > --Hannes > > > On 4/14/11, Edgar J. De Cleene <[hidden email]> wrote: >> >> >> >> On 4/14/11 8:09 AM, "Bert Freudenberg" <[hidden email]> wrote: >> >>> No, there are problems with your code. See my other message in the >>> "Monticello-edc.456" thread. >>> >>> - Bert - >> >> Well, now i know this is wrong. >> Like know what is right. >> >> Usinhg DependencyBrowser in fresh image >> >> MCTool>>#buttonRow: We have AlignmentMorph and PluggableButtonMorph >> MCWorkingCopyBrowser>>#editLoadScripts We have MenuMorph >> MCTool>>#multiListMorph:selection:listSelection:menu: We have >> PluggableListMorphOfMany >> >> And we could follow, but the critics was for >> MCRepository class>>#fillInTheBlankConfigure: >> >> I copy the code >> >> fillInTheBlankConfigure: aTemplateString >> | chunk repo | >> >> aTemplateString ifNil: [ ^ false ]. >> chunk := FillInTheBlankMorph >> request: self fillInTheBlankRequest >> initialAnswer: aTemplateString >> centerAt: Sensor cursorPoint >> inWorld: World >> onCancelReturn: nil >> acceptOnCR: false >> answerExtent: 400@120. >> >> chunk >> ifNotNil: [ >> repo := self readFrom: chunk readStream. >> repo creationTemplate: chunk. >> ]. >> >> ^ repo >> >> >> Is this best as my solution? >> Was in place for four years .... stephaneducasse 2/4/2006 20:47 >> >> Edgar >> >> >> >> which shows DependencyBrowser working in SimpleMorphic and selecting Morphic dependencies I think i get all complains and could do buildWith: in all places following as example code of Andreas in FileContentsBrowser>>#createViews Edgar |
Edgar
That screen shot http://tech.groups.yahoo.com/group/squeak/message/161102 says that the Morphic package depends on Monticello. The dependency created by the classes ListItemWrapper MenuMorph PluggableListMorph PluggableListMorph.. PluggableMultiColumn... SimpleHierarchicalLis... SystemWindow TheWorldMenu (I just used the screen shot and not all class names were fully visible) So what does this mean for integrating SimpleMorphic? --Hannes On 4/14/11, Edgar J. De Cleene <[hidden email]> wrote: > > > > El 14/04/11 12:28, "Hannes Hirzel" <[hidden email]> escribió: > >> Edgar >> >> >> Do you mean what is shown in the screen shot? >> >> --Hannes >> >> >> On 4/14/11, Edgar J. De Cleene <[hidden email]> wrote: >>> >>> >>> >>> On 4/14/11 8:09 AM, "Bert Freudenberg" <[hidden email]> wrote: >>> >>>> No, there are problems with your code. See my other message in the >>>> "Monticello-edc.456" thread. >>>> >>>> - Bert - >>> >>> Well, now i know this is wrong. >>> Like know what is right. >>> >>> Usinhg DependencyBrowser in fresh image >>> >>> MCTool>>#buttonRow: We have AlignmentMorph and PluggableButtonMorph >>> MCWorkingCopyBrowser>>#editLoadScripts We have MenuMorph >>> MCTool>>#multiListMorph:selection:listSelection:menu: We have >>> PluggableListMorphOfMany >>> >>> And we could follow, but the critics was for >>> MCRepository class>>#fillInTheBlankConfigure: >>> >>> I copy the code >>> >>> fillInTheBlankConfigure: aTemplateString >>> | chunk repo | >>> >>> aTemplateString ifNil: [ ^ false ]. >>> chunk := FillInTheBlankMorph >>> request: self fillInTheBlankRequest >>> initialAnswer: aTemplateString >>> centerAt: Sensor cursorPoint >>> inWorld: World >>> onCancelReturn: nil >>> acceptOnCR: false >>> answerExtent: 400@120. >>> >>> chunk >>> ifNotNil: [ >>> repo := self readFrom: chunk readStream. >>> repo creationTemplate: chunk. >>> ]. >>> >>> ^ repo >>> >>> >>> Is this best as my solution? >>> Was in place for four years .... stephaneducasse 2/4/2006 20:47 >>> >>> Edgar >>> >>> >>> >>> > See attached to http://tech.groups.yahoo.com/group/squeak/message/161102 > which shows DependencyBrowser working in SimpleMorphic and selecting Morphic > dependencies > > I think i get all complains and could do buildWith: in all places following > as example code of Andreas in FileContentsBrowser>>#createViews > > Edgar > > > > |
On 4/17/11 1:36 PM, "Hannes Hirzel" <[hidden email]> wrote: > Edgar > > That screen shot http://tech.groups.yahoo.com/group/squeak/message/161102 > says > > that the Morphic package depends on Monticello. > > The dependency created by the classes > > ListItemWrapper > MenuMorph > PluggableListMorph > PluggableListMorph.. > PluggableMultiColumn... > SimpleHierarchicalLis... > SystemWindow > TheWorldMenu > > (I just used the screen shot and not all class names were fully visible) > > So what does this mean for integrating SimpleMorphic? > > --Hannes Means we should change all call references to Morph(s) in Monticello with some best. All references to FillInTheBlankMorph could easily be changed to UIManager default. But we do not have this for MenuMorph, etc. One idea I send some days ago could be use ToolBuilder default and his subclasses where we have a category widget classes with methods like buttonClass, listByItemClass, menuClass, etc. And still we must add some code for solve all .... As example other mcz I send for review have ----- Method: MCTool>>textMorph: (in category 'morphic ui') ----- textMorph: aSymbol + ^ UIManager default toolBuilder textPaneClass on: self text: aSymbol accept: (aSymbol, ':') asSymbol! * ^ PluggableTextMorph on: self text: aSymbol accept: (aSymbol, ':') asSymbol! So i wait in the bench waiting our coach say this is the way to go or not ... Edgar |
Edgar
thank you for the added info and the proposal. Before we go further into this, let me ask first. In which way is it a problem that Morphic depends on Monticello? --Hannes On 4/18/11, Edgar J. De Cleene <[hidden email]> wrote: > > > > On 4/17/11 1:36 PM, "Hannes Hirzel" <[hidden email]> wrote: > >> Edgar >> >> That screen shot http://tech.groups.yahoo.com/group/squeak/message/161102 >> says >> >> that the Morphic package depends on Monticello. >> >> The dependency created by the classes >> >> ListItemWrapper >> MenuMorph >> PluggableListMorph >> PluggableListMorph.. >> PluggableMultiColumn... >> SimpleHierarchicalLis... >> SystemWindow >> TheWorldMenu >> >> (I just used the screen shot and not all class names were fully visible) >> >> So what does this mean for integrating SimpleMorphic? >> >> --Hannes > > Means we should change all call references to Morph(s) in Monticello with > some best. > > All references to FillInTheBlankMorph could easily be changed to UIManager > default. > > But we do not have this for MenuMorph, etc. > One idea I send some days ago could be use ToolBuilder default and his > subclasses where we have a category widget classes with methods like > buttonClass, listByItemClass, menuClass, etc. > And still we must add some code for solve all .... > > As example other mcz I send for review have > > > ----- Method: MCTool>>textMorph: (in category 'morphic ui') ----- > textMorph: aSymbol > + ^ UIManager default toolBuilder textPaneClass on: self text: aSymbol > accept: (aSymbol, ':') asSymbol! > * ^ PluggableTextMorph on: self text: aSymbol accept: (aSymbol, ':') > asSymbol! > > So i wait in the bench waiting our coach say this is the way to go or not > ... > > Edgar > > > > |
On 4/18/11 8:02 AM, "Hannes Hirzel" <[hidden email]> wrote: > Edgar > > thank you for the added info and the proposal. > > Before we go further into this, let me ask first. > > In which way is it a problem that Morphic depends on Monticello? Monticello depends on Morphic as his code is today.. So, if in the future you wish get rid of Morphic and unload it and instead use other UI like SimpleMorphic , you must change all references to Morphic classes like MeNuMorph, AlignmentMorph, etc. Edgar |
On Mon, 18 Apr 2011, Edgar J. De Cleene wrote:
> > > > On 4/18/11 8:02 AM, "Hannes Hirzel" <[hidden email]> wrote: > >> Edgar >> >> thank you for the added info and the proposal. >> >> Before we go further into this, let me ask first. >> >> In which way is it a problem that Morphic depends on Monticello? > > Monticello depends on Morphic as his code is today.. > So, if in the future you wish get rid of Morphic and unload it and instead > use other UI like SimpleMorphic , you must change all references to Morphic > classes like MeNuMorph, AlignmentMorph, etc. Bert mentioned that MC was fully toolbuilderized, so this problem was solved. We should check that solution first IMHO. Levente > > Edgar > > > > |
I agree, where do we start checking?
--Hannes On 4/18/11, Levente Uzonyi <[hidden email]> wrote: > On Mon, 18 Apr 2011, Edgar J. De Cleene wrote: > >> >> >> >> On 4/18/11 8:02 AM, "Hannes Hirzel" <[hidden email]> wrote: >> >>> Edgar >>> >>> thank you for the added info and the proposal. >>> >>> Before we go further into this, let me ask first. >>> >>> In which way is it a problem that Morphic depends on Monticello? >> >> Monticello depends on Morphic as his code is today.. >> So, if in the future you wish get rid of Morphic and unload it and instead >> use other UI like SimpleMorphic , you must change all references to >> Morphic >> classes like MeNuMorph, AlignmentMorph, etc. > > Bert mentioned that MC was fully toolbuilderized, so this problem was > solved. We should check that solution first IMHO. > > > Levente > >> >> Edgar >> >> >> >> > > |
Free forum by Nabble | Edit this page |